
275 
Number of globules in Relative 
Date. .0001 cubic millimeter. size. 
May 18, evening haf 239 
May 19, morning 144 268 
May 21, morning 124 357 
May 22, evening 151 296 
May 23, morning 150 305 
May 25, morning eee bot Se 
May 25, evening i 148 304 
May 27, morning 149 306 
May 27, evening 148 307 
With the exception of Emm, the globules of whose milk have al- 
ways been very large, the numbers representing the relative size of 
the globules, represent quite closely, so far as the cows have been 
tested, their relative value for butter production. The cows, Hilda 
D, Evelina of Verna, and Edessa, gave in an official test 21 Ibs. 2 OZS., 
19 lbs. 10 ozs. and 10 Ibs. 6 ozs. of butter, respectively, in 7 days. 
(See Jersey tests in this report.) Jem has made 15 lbs. of butter in 
seven days. So far as the cows of the Station herd have been tested, 
the native cow, Ceres, stands next to Jem in per cent. of fat recovered 
in her butter; then comes Meg, and last of all, Juno. The COWS, 
Emm and Ann, have not received sufficient test to indicate their 
standing. 
There seems also to be some relation between the melting points of 
the butter fats and the size of the globules (see table above), and al- 
though these are not directly proportional they are sufficiently 
marked to indicate some dependence upon each other. 
An inspection of results obtained with Juno and Ceres during the 
feeding experiments also show decided changes which may and may 
not be due to the ration. All discussion of this question has been 
intentionally postponed until further and more conclusive data are 
obtained. 
Butrer TEstTs. 
The butter tests of the past year have, with the exception of a 
short test of the Jersey cow, Jem, been made with the native cows, 
Ceres and Juno. The tests were supplementary to the feeding ex- 
periments with starch refuse and glucose meal, and the results ob- 
tained will be found in the report of the director on these foods. 
There are, however, some relations shown by the tests that are inde- 
pendent of the foods, and which may properly be discussed here. 
The trials were all made with the Cherry test churn described in 
the report of this Station for 1884. The quantity of milk taken was 
in every instance 400 grams (about a pint); this was set in glass stop- 
pered quart bottles in a tank through which water from the spring 
was running constantly. The temperature of this water varied from 
about 50 F. at the beginning of the experiments in December to 40 
F.in Feb. and March. ‘The time set was usually 54 hours, but in a 
few cases was only 30, and in a few more was extended to 80. The 
best results were obtained when the milk was from 48 to 60 hours 
old. Before churning, the bottles were removed from the water and 
placed upon a table in a warm room for several hours, and, when 
ready to churn, the temperature was adjusted by placing the bottles 
in warm or cold water, as appeared necessary. ‘The temperature of 
